First Author

Despite great demand for nanoscale sensors by the security, environmental and healthcare industries, the performance of current devices is disappointing owing to technical limitations. The trickiest job is connecting nanometre-sized structures to the much larger silicon wafers required to produce functioning devices. Integrating different materials, such as carbon- and silicon-based structures, has also proved problematic. And device performance is often sacrificed when bulk silicon is etched down to the nanometre scale. When Yale University graduate student Eric Stern joined a project — funded by the US Defense Advanced Research Projects Agency — to build such a nanosensor, he assessed the limitations of existing nanomaterials. His modification of a conventional silicon-based approach led to a new nanowire design for an immunodetection device (see page 519).

What was the project's original goal?

At the outset, the government had grave concerns about stockpiled chemical and biological weapons in Iraq. The goal was to develop a general sensor that could detect these agents and be worn by every soldier. Although such a device is still very much a work in progress, we've hopefully moved it a step closer to reality.

Did you stumble on any unexpected applications for the nanosensors?

The most exciting application is in detecting stimulus-induced cellular responses. Sensors with this capability could be used in diagnostics to differentiate between various potentially pathogenic cell types. They would require samples of only a few cells and no artificial labels.

What was the main hurdle to overcome?

After much study of thousands of nanotube and nanowire devices, we decided to use a more conventional approach and etch the nanowires from bulk silicon. The real trick was finding the best way to do this in order to get pristine nanostructures with smooth side walls. From a sensing-device point of view, it's never been done before.

What's been the biggest victory for you?

I was frustrated with the inability of groups to reproduce high-quality semiconducting nanowires required for sensing studies. This limitation has prevented nanosensing from taking off. Our approach lays some of the groundwork necessary to overcome this.

Your team represents five different departments at Yale. Was that by design?

Yes. A strong collaboration between specialists from multiple disciplines must be created to perform a study like this.